Finding natural clusters having minimum description length
- 4 December 2002
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. i, 438-442
- https://doi.org/10.1109/icpr.1990.118142
Abstract
A two-step procedure that finds natural clusters in geometric point data is described. The first step computes a hierarchical cluster tree minimizing an entropy objective function. The second step recursively explores the tree for a level clustering having minimum description length. Together, these two steps find natural clusters without requiring a user to specify threshold parameters or so-called magic numbers. In particular, the method automatically determines the number of clusters in the input data. The first step exploits a new hierarchical clustering procedure called numerical iterative hierarchical clustering (NIHC). The output of NIHC is a cluster tree. The second step in the procedure searches the tree for a minimum-description-length (MDL) level clustering. The MDL formulation, equivalent to maximizing the posterior probability, is suited to the clustering problem because it defines a natural prior distribution.Keywords
This publication has 5 references indexed in Scilit:
- Attributed image matching using a minimum representation size criterionPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- Constructing simple stable descriptions for image partitioningInternational Journal of Computer Vision, 1989
- A Universal Prior for Integers and Estimation by Minimum Description LengthThe Annals of Statistics, 1983
- An information measure for single link classificationThe Computer Journal, 1975
- An Information Measure for ClassificationThe Computer Journal, 1968